Journal of Neuroscience Research 31:494-501 (1992)

Cytoskeletal Elements Regulate the Distribution of Nerve Growth Factor Receptors in PC12 Cells P.E. Spoerri and F.J. Roisen Department of Anatomical Sciences and Neurobiology , University of Louisville School of Medicine, Louisville, Kentucky

Nerve growth factor receptor (NGFR)-like immunoreactivity (IR) was studied in PC12 cells treated for 96 hr with NGF (40 ng/ml), using immunogold labeling and electron microscopic morphometric analysis. The cells were exposed to the anti-NGFR antibody 192-IgG, followed by immunoglobulin (IgG) conjugated with colloidal gold. PC12 cells exhibited occasional gold label (positive NGFR-IR) on all surfaces. Cells treated with colcemid (0.05 pg/ml) or cytochalasin D (2 pg/ml), which limit microtubule (MT) and microfilament (MF) formation, respectively, displayed an increased NGFR-IR in terms of gold labeling. NGFR-IR was also seen on taxol (0.85 pg/ml)exposed cells, an agent that promotes MT assembly. Cells treated simultaneously with cytochalasin D and taxol had a dramatically augmented NGFR-IR on their surfaces, which exceeded levels obtained with either agent alone. Prominent NGFR-IR was localized frequently in coated endocytotic vesicles, in smooth endoplasmic reticulum, and in secondary multivesicular lysosomes, in both treated and untreated cells. The results suggest that a large number of NGFRs (positive NGFR-IR) in PC12 cells are cryptic and not available for ligand binding. Changes in cytoskeletal organization that may affect mobility of integral membrane proteins can modulate the distribution of NGFR-IR on neuronal surfaces. Key words: NGF receptor, immunolocalization, colloidal gold, ultrastructure

Kreutzberg, 1987). The development of a rat monoclonal anti-NGFR antibody, 192-IgG (Chandler et al., 1984), has allowed the immunocytochemical demonstration of the receptor protein in rat central nervous system neurons (Pioro et al., 1990). There is biochemical evidence that 192-IgG (Chandler et al., 1984) identifies not only the low-affinity form of the NGFR, but a protein common to both low- and high-affinity forms of the NGFR (Green and Greene, 1986). Cross linking [‘251]NGFto its receptor in C6 rat glioma cells followed by immunoprecipitation with 192-IgG and sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) revealed a major band of 100 kD and a minor band of about 200 kD corresponding to the NGF/NGFR complex (Kumar et al., 1990), similar to values reported for PC12 cells (Green and Greene, 1986). In the present study, the reaction product with anti-NGFR antibody should be considered to represent “NGFR-like” immunoreactivity, as suggested by Pioro et al. (1990). Except for transmission electron microscopic (TEM) studies localizing internalized [‘251]NGF in cultured PC12 cells (Hogue-Angelettiet al., 1982; Rohrer et al., 1982; Bernd and Greene, 1983) and in peripheral sympathetic neurons (Claude et al., 1982), there have been no immunolocalization studies at the TEM level demonstrating sites of NGFR in PC12 cells. Therefore, this study was undertaken using 192-IgG to identify the surface and subcellular distribution of NGFR-IR in these cells. In addition, since some NGF becomes associated with cytoskeletal elements (Schechter and Bothwell, 1981; Vale et al., 1985) and NGF binding sites have been

INTRODUCTION The clonal rat pheochromocytoma PC12 cell responds to nerve growth factor (NGF) by cessation of division and development of a number of properties characteristic of mature sympathetic neurons (Green et al., 1986). It is believed that the neuronotrophic-neuritogenic effects of NGF are mediated via binding to the nerve growth factor receptor (NGFR), specifically its high-affinity form (Richardson et al., 1986; Raivich and 0 1992 Wiley-Liss, Inc.

Received June 4, 1991; revised September 23, 1991; accepted September 30, 1991. Dr. P.E. Spoerri’s present address is Department of Molecular Neurobiology , Fidia Research Laboratories, Via Ponte della Fabbrica 3/A, 35031-Abano Terme (PD), Italy. Address reprint requests to Dr. F.J. Roisen, Department of Anatomical Sciences and Neurobiology, University of Louisville School of Medicine, Louisville, KY 40292.

Cytoskeletal Alterations Affect NGF Receptor demonstrated on microtubules (MTs) and microfilaments (MFs) (Nasi et al., 1982), cytoskeletal altering agents were applied prior to immunolocalization of the NGFR, to determine the degree and nature of the NGFR associations to the underlying cytoskeleton.

MATERIALS AND METHODS Materials RPMI 1640 medium, fetal bovine serum (FBS), horse serum, Hanks’ balanced salt solution (HBSS), and colcemid were obtained from Gibco (Grand Island, NY), cytochalasin D from Sigma (St. Louis, MO), taxol from Natural Products (branch of the National Cancer Institute, Rockville, MD), anti-NGF receptor antibody from Boehringer Mannheim (Indianapolis, IN), goat antimouse (GAM) affinity-purified antibody conjugated with 10 nm gold particles from Janssen (Piscataway, NJ), Epon 8 12 and glutaraldehyde from Electron Microscopy Sciences (Washington, PA), and OsO, from Degussa Corporation (Teterboro, NJ). Cell Culture Rat pheochromocytoma PC12 cells (a gift of Dr. M.A. Bothwell) were maintained as stocks in RPMI 1640 medium supplemented with 10% horse serum and 5% FBS in a humidified atmosphere of 5% CO, and 95% air at 36°C. Cells were dislodged mechanically from the flask and passed every 5 days. Cells for immunoelectron microscopy were plated on 2-mm collagen-coated glass coverslips at a density of 2.0 X lo4 cells per coverslip. NGF (a gift of Dr. R. Murphy) was added at a concentration of 40 ng/ml, sufficient to produce reliable neurite outgrowth within 96 hr without obscuring further stimulation by other test agents. Cytoskeletal-altering agents were introduced prior to fixation for immunoelectron microscopy at time points previously shown to have distinct ultrastructural effects while preserving adequate neurite length (Spero and Roisen, 1985a,b). Colcemid (0.05 pg/ml) was added to cultures 20 min prior to fixation. Cytochalasin D (2 pg/ ml) and/or taxol(0.85 pg/ml) were added 2 hr or 30 min prior to fixation, respectively. Immunoelectron Microscopy PC12 cells grown under the above conditions were washed once with HBSS. One hundred microliters of fresh medium containing 1: 10 dilution of monoclonal antibody to rat nerve growth factor receptor anti-NGFR; 192-IgG; Chandler et al., 1984) was added to coverslips and incubated with gentle agitation for 30 min at 36°C in a humidified atmosphere of 5% CO, and 95% air. After removal of the medium, cells were washed twice with HBSS, and fresh culture medium containing a 1:15 di-

495

lution of goat antimouse (GAM) antibody conjugated with 10 nm gold particles was added with gentle agitation as described above. Coverslips were washed five times with HBSS and processed for TEM.

TEM For TEM, cells grown on collagen-coated glass coverslips were drained of medium, fixed at room temperature in 2.5% glutaraldehyde in 0.08 M phosphate buffer (pH 7.3) for 30 min, and postfixed in 1% OsO, in 0.08 M phosphate buffer containing 0.1 M sucrose for 60 min. After dehydration and embedding, selected areas were sectioned and stained with a saturated aqueous solution of uranyl acetate and examined and photographed using a Philips CMlO electron microscope. Morphometric Analysis Using the TEM conditions described in this study, a minimum of 50 electron micrographs were obtained from randomly selected samples of triplicate cultures of the various treatment groups. From electron micrographs, (magnification X 48,000),data were extracted using direct planimetry measurements (Bendayan, 1984). After evaluation of the distance of surface (Sa) occupied by the plasmalemma in a defined compartment using Sigma Scan, the number of gold particles (Ni) present per unit arealelectron micrograph was counted and the density of labeling (Ns) calculated: Ns = Ni/Sa. Statistical Analysis One-way ANOVAs were used to test the main effect of treatment (colcemid, cytochalasin D, taxol, cytochalasin D + taxol). Where significant main effects were found, Fisher’s posthoc test was employed to evaluate differences between groups. Statistical significance was defined as a value of P < 0.05. RESULTS Ultrastructural immunolocalization of the NGFR revealed that PC 12 cells not exposed to cytoskeletal altering agents exhibited occasional gold labeling (positive NGFR-IR) on somatic and neuritic surfaces (Fig. la,b). Exposure of cells to colcemid or cytochalasin D prior to anti-NGFR treatment dramatically increased gold labeling on all surfaces (Fig. lc-f). Similarly, exposure of cells to taxol also augmented the gold labeling (Fig. lg,h). Simultaneous treatment with cytochalasin D and taxol further potentiated the enhancement of gold labeling (positive NGFR-IR) on somata and neuritic surfaces (Fig. 1i,j). Morphometric analysis of electron micrographs revealed a significant increase in gold particles (positive NGFR-IR) in the treated groups compared with the con-

496

Spoerri and Roisen

Fig. 1. Electron micrographs of differentiated PC12 cells in medium supplemented with NGF (40 ng/ml) for 96 hr and exposed to anti-NGFR (192-IgG) followed by GAM-IgG conjugated with 10 nm gold particles. a,b: Control cells exhibiting occasional NGFR-IR on their surfaces. c,d: Colcemid (0.05 pg/ml)-treated cells (to disrupt MTs). e,f: Cytochalasin D (2p,g/ml)-treatedcells (to disrupt MFs). g,h: Taxol (0.85pg/

m1)-treated cells (to increase MTs). i j : Cytochalasin D (2 pg/ml) + taxol (0.85 pg/ml)-treated cells. The cytoskeletal altering agents increased the level of gold label (positive NGFR-IR) on all exposed surfaces. Maximal NGFR-IR was obtained by simultaneous treatment with cytochalasin D + taxol. X45,OOO.

trol group. The differences in NGFR-IR between the treated groups were also significant (Fig. 2). Immunoreactive gold particles were internalized in both control and treated (colcemid, cytochalasin D, taxol, cytochalasin D taxol) PC12 cells (Fig. 3a-h). The NGFR-immunoreactive gold particles were found engulfed in endocytotic coated vesicles of the plasmalemma. Similarly, gold-containing coated vesicles were frequently observed within the cytoplasm. The immunoreactive gold particles were also seen to be localized within the smooth endoplasmic reticulum (SER) and within secondary lysosomes (multivesicular appearance) found in the perinuclear region and in neuritic shafts in control and treated cells (Fig. 3a-h). No association between gold particles and MTs or MFs was ever observed. Disruption of MTs or MFs with colcemid or cytochalasin D, respectively, or stabilization of MTs with taxol did not affect the degree of internalization of immunoreactive gold particles. Internalization of immunoreactive NGFR occurred in both

control PC12 cells and in cells treated with cytoskeletal altering agents.

+

DISCUSSION PC 12 cells grown in the presence of NGF exhibited occasional NGFR-IR. The immunolocalization of NGFR with colloidal gold revealed short, nonrandom arrangements of gold particles on the plasmalemma, suggesting that NGFR may be associated with the underlying cytoskeleton. Treatment of the cells with colcemid followed by immunolocalization with anti-NGFR antibody (192-IgG) did, in fact, increase NGFR-IR significantly in terms of gold labeling compared to the untreated cells. Colcemid binds tubulin subunits and inhibits microtubule polymerization (Luduena, 1979). Cells exposed to cytochalasin D also showed increased NGFR-IR. Cytochalasins are potent inhibitors of actin polymerization and, among these, cytochalasin D is the most specific inhibitor of MF assembly (Yahara et al., 1982).

Fig. lc-f.

498

Spoerri and Roisen

Exposure of NGF-pretreated PC12 cells to taxol also enhanced NGFR-IR on all surfaces. This enhancement was less than that seen in cultures treated with either colcemid or cytochalasin D. Taxol binds tubulin

and has been shown to lower the tubulin concentration required for microtubule assembly, thereby promoting microtubule polymerization. This may reduce the ability of tubulin to bind to the cell surface (Parness and Hor-

Cytoskeletal Alterations Affect NGF Receptor

161

CONTROL COLCEMID C M O D

TAXOL

CMO D

-+ TAXOL

Fig. 2. Number of gold particles (positiveNGFR-IR) per unit cell surface area. Results are expressed as means -+ SEM of the number of gold particles in a minimum of 50 electron micrographs per treatment. These were obtained from random samples of triplicate cultures from three independent experiments. One-way ANOVAs were used to analyze the effect of treatment, which was significant. (F = 103, 26; d.f. = 4, 249; P < 0.001). Fischer's posthoc test was employed to evaluate differences between individual groups. There was an increase in gold particles (positiveNGFR-IR) in each treated group over that seen in controls ( P < 0.05). In addition, the differences between treatment groups were significant ( P < 0.05).

499

tribution of NGFR. Previous studies described NGF associations with cytoskeletal elements following its binding to receptors on PC12 cells (Schechter and Bothwell, 1981; Nasi et al., 1982; Vale et al., 1985). Several studies have reported on the fate of ['251]NGFfollowing its binding to receptors on the surface of PC12 cells. This ligand has been localized in SER, lysosomes, and the nuclear membrane (Hogue-Angeletti et al., 1982; Bernd and Greene, 1983). Others found no evidence of association with the nucleus (Hogue-Angeletti et al., 1982; Rohrer et al., 1982). Recently, NGFR-immunoreactive central nervous system neurons of the nucleus basalis magnocellularis of the adult rat were examined with TEM. NGFR-IR was consistently present along the plasmalemma and frequently in intracellular sites, which included the rough endoplasmic reticulum (RER), Golgi apparatus, secondary lysosomes, and nuclear membrane (Pioro et al., 1990). A thorough examination of a large number of sections of PC12 cells labeled with the anti-NGFR antibody 192IgG gave no evidence of association of the NGF-R (positive NGF-R-IR) with the RER, the Golgi apparatus or the nuclear membrane. Gold particles were, however, internalized in endocytotic-coated vesicles. Prominant NGFR-IR was frequently localized in the SER and in secondary lysosomes in untreated PC12 cells or in cells treated with cytoskeletal altering agents. PC12 cells exposed to colcemid, cytochalasin D, and/or taxol exhibited an internalization of NGFR immunoreactive-like gold particles that did not differ between treated and untreated cultures. Positive immunoreactivity was seen in the secondary lysosomes of all PC12 cells examined. Neither MTs nor MFs were involved in the transport pathways leading to NGF degradation, in contrast to the findings of Kasaian and Neet (1988), suggesting that these structures may be involved in this pathway. Our ultrastructural evidence of NGFR internalization indicates that the antibody 192-IgG reacting with IgG-conjugated gold also reveals the high-affinity form of the receptor (presumably in the copresence of the lowaffinity component), since the low-affinity form of this receptor by itself is not internalized (Hosang and Shooter, 1987; Pioro et al., 1990). Surface NGF receptors, as detected by NGFR-like immunoreactivity, appear to be under the direct influence of MTs and MFs, and changes in the cytoskeletal organization may function to modulate the distribution of the NGFR.

witz, 1981). Cells treated simultaneously with taxol and cytochalasin D exhibited the most dramatic change in the distribution of NGFR-IR in terms of increased gold labeling compared with untreated cells. Taxol, which may reduce the ability of tubulin to bind to the cell surface, may also strengthen the association of actin filaments from membranous components, as previously suggested (Spero and Roisen, 1985a,b), and may facilitate a possible association between microtubules and the plasma membrane. Disturbances in membrane dynamics caused by these two agents could explain the increased appearance of previously cryptic NGF-binding sites (positive NGFR-IR) on the plasmalemma as determined by analytical TEM morphometry. NGFRs are kinetically heterogeneous populations (Henup and Thoenen, 1979), which, in the absence of NGF, can exist in two states, exposed and hidden. Only the exposed receptors can bind NGF (Cattaneo et al., 1983). Once NGF is bound to exposed receptors, it may participate in the fate of the receptor and become cryptic itself. Cytoskeletal altering agents may affect the mobility of integral membrane proteins and expose cryptic ACKNOWLEDGMENTS receptors on the surface. These receptors may be located The authors thank Dr. Stephen D. Skaper for critclose to the plasma membrane, linked to cytoskeletal organelles, folded in the membrane, or clustered in a ical reading of the manuscript, Dr. William E. Renehan manner that precludes labeling. Changes in cytoskeletal for use of the Sigma Scan, Dr. Laura Schweitzer for help organization (affecting MTs and MFs) modulate the dis- with the statistical analysis, and Ms. Cathie G. Caple and

Fig. 3. Profiles from representative NGFR-immunoreactive PC12 cells revealing the presence of reaction product (gold particles) beneath the plasmalemma or in the perinuclear region. a: NGFR-IR is seen within endocytotic-coated vesicles (CV) and in the smooth endoplasmic reticulum (SER) at the periphery of control (untreated) cells. b: Gold particles (positive NGFR-IR) from control cells are also found in secondary lysosomes (LY). c,d: Colcemid-treated cells reveal NGFR-IR

within CVs, SER, and secondary LYs. e,f: Similarly, cytochalasin D-treated cells have NGFR-IR gold particles within SER, CVs (not shown), and secondary LYs. g,h: Taxol alone or in combination with cytochalasin D (g and h, respectively) also induced active internalization of the NGFR-IR gold particles. The latter are seen in CVs, SERs, and secondary LYs. X 45,000.

Cytoskeletal Alterations Affect NGF Receptor

Ms. Glee Yorke for technical assistance. In addition, thanks are due Ms. Antonia Bedeschi for typing the manuscript. This work was supported by USPHS grants NS24524 and DE07734.

REFERENCES Bendayan M (1984): Protein A-gold electron microscopic immunocytochemistry: methods, applications, and limitations. J Electron Microsc Techno1 1:243-270. Bernd P, Greene LA (1983): Electron microscopic radioautographic localization of iodinated nerve growth factor bound to and internalized by PC12 cells. J Neurosci 3:631-643. Cattaneo A, Biocca S , Nasi S , Calissano P (1983): Hidden receptors for nerve growth factor in PC12 cells. Eur J Biochem 135: 285 -290. Chandler CE, Parsons LM, Hosang M, Shooter EM (1984): A monoclonal antibody modulates the interaction of nerve growth factor with PC12 cells. J Biol Chem 259:6882-6889. Claude P, Hawrot E, Dunis DA, Campenot RB’(1982): Binding, internalization, and retrograde transport of ‘251-nerve growth factor in cultured rat sympathetic neurons. J Neurosci 2:431442. Green SH, Greene LA (1986): A single Mr = 103, OOO ‘251-p-nerve growth-factor-affinity labeled species represents both the low and high affinity forms of the nerve growth factor receptor. J Biol Chem 261 :15316-15326. Green SH, Rydel RE, Connolly JL, Greene LA (1986): PC12 cell mutants that possess low- but not high-affinity nerve growth factor receptors neither respond to nor internalize nerve growth factor. J Cell Biol 102:830-843. Hermp K, Thoenen H (1979): Properties of the nerve growth factor receptor of a clonal line of rat pheochromocytoma (PC12) cells. Exp Cell Res 121:71-78. Hogue-Angeletti R, Stieber A, Gonatas NK (1982): Endocytosis of nerve growth factor by PC12 cells studied by quantitative ultrastructural autoradiography. Brain Res 241: 145-156. Hosang M, Shooter EM (1987): The internalization of nerve growth factor by high-affinity receptors on pheochromocytoma PC 12 cells. EMBO J 6:1197-1202. Kasaian MT, Neet KE (1988): Internalization of nerve growth factor by PC12 cells. A description of cellular pools. J Biol Chem 26315083-5090.

501

Kumar S, Huber J , Pena LA, Perez-Polo JR, Werrbach-Perez K, de Vellis J (1990): Characterization of functional nerve growth factor receptors in a CNS glial cell line: monoclonal antibody 217c recognizes the nerve growth factor receptor on C6 glioma cells. J Neurosci Res 27”-417. Luduena RF (1979): Biochemistry of tubulin. In Roberts K, Hyams JS (eds): “Microtubules.” New York: Academic Press, pp 1-66. Nasi S , Cirillo D, Naldini L, Marchisio PC, Calissano P (1982): Microtubules and microfilaments in fixed and permeabilized cells are selectively decorated by nerve growth factor. Proc Natl Acad Sci USA 793320-824. Parness J, Horwitz SB (1981): Taxol binds to polymerized tubulin in vitro. J Cell Biol 91:479-487. Pioro EP, Ribeiro-da-Silva A, Cuello AC (1990): Immunoelectron microscopic evidence of nerve growth factor receptor metabolism and internalization in rat nucleus basalis neurons. Brain Res 527:109-115. Raivich G, Kreutzberg GW (1987): The localization and distribution of high affinity p-nerve growth factor binding sites in the central nervous system of the adult rat. A light microscopic autoradiographic study using [‘251]p-nerve growth factor. Neuroscience 20:23-36. Richardson PM, Verge Issa VMK, Riopelle RJ (1986): Distribution of neuronal receptors for nerve growth factor in the rat. J Neurosci 6:23 12-2321. Rohrer H, Schafer T, Korsching S , Thoenen H (1982): Internalization of nerve growth factor by pheochromocytoma PC12 cells: Absence of transfer to the nucleus. J Neurosci 2:687-697. Schechter AL, Bothwell MA (1981): Nerve growth factor receptors on PC12 cells: Evidence for two receptor classes with differing cytoskeletal association. Cell 245367474. Spero DA, Roisen FJ (1985a): Neuro-2a neuroblastoma cells form neurites in the presence of taxol and cytochalasin D. Dev Brain Res 23:155-159. Spero DA, Roisen FJ (1985b): Gangliosides induce microfilamentdependent changes in membrane surface activity of Neuro-2a neuroblastoma cells. Int J Dev Neurosci 3:631-642. Vale RD, Ignatius MJ, Shooter EM (1985): Association of nerve growth factor receptors with the Triton X-100 cytoskeleton of PC12 cells. J Neurosci 92762-2770. Yahara I, Harada F, Sekita S , Yoshihira K, Natori S (1982): Correlation between effects of 24 different cytochalasins on cellular structures and cellular events and those on actin in vitro. J Cell Biol 92:69-78.